Abstract
Closed incision negative pressure wound therapy (CINPWT) has been shown to be clinically effective compared with the traditional gauze dressing, reducing surgical site infections and wound complications. We evaluated the effect of CINPWT compared with gauze dressing on the need for revision surgery and survival after non‐traumatic major lower amputation. We included 309 patients undergoing 403 major lower amputations in a retrospective study from January 1, 2010 to November 23, 2017. A total of 139 patients received CINPWT, and 170 patients received stump bandage. There was no statistically significant difference between the two groups regarding the need for revision surgery (P = .45). Fourteen stump bandage patients and 15 CINPWT patients died in hospital (P = .57). One year after amputation, 55 CINPWT patients and 66 stump bandage patients had died (P = .82). Survival probabilities adjusted for age and gender 2 years after amputation were .52 (.43‐.61) and .49 (.42‐.58), respectively, and 3 years after amputation were .36 (.25‐.50) and .39 (.32‐.47), respectively. We also found no significant difference in the need for revision surgery in survival probabilities up till 3 years after amputation between patients treated with CINPWT and patients treated with gauze bandage postoperatively.
Keywords: amputation, diabetes, infection
1. INTRODUCTION
Compared with traditional gauze dressing, closed incision negative pressure wound therapy (CINPWT) has been shown to be clinically effective in the treatment of high‐risk wounds after severe skeletal trauma and in arthroplasty revision surgery.1, 2, 3 In 2017, Willy et al. published consensus recommendations for the use of CINPWT, describing patient characteristics, incisions, and operations with a high risk of developing surgical site infections and wound complications. One of the operation‐related risk factors mentioned in the article was above‐ and below‐knee amputations.4
However, even before these recommendations, CINPWT was widely used in multi‐morbid patients undergoing major lower non‐traumatic (above‐ankle) amputation, although there was no evidence of this treatment.5, 6 Today, specific indications for use of CINPWT continue to be defined.
Advanced age, male gender, multi‐morbidity, and polypharmacy characterise patients with non‐traumatic major lower‐extremity amputations. They suffer from long‐term complications of atherosclerosis and diabetes resulting in ischaemia and infected and necrotic wounds.7, 8 In Denmark, 30‐day mortality has been reported to be 16% to 35% and 1‐year mortality to be 37% to 59%9, 10, 11 in this frail group of patients.
The purpose of this study was to evaluate the effect of CINPWT compared with gauze dressing on survival after non‐traumatic major lower amputation. Our hypothesis was that the clinical outcomes did not depend on bandage type. We analysed the use of revision surgery; subsequent amputation; length of stay in hospital; in‐hospital mortality; and mortality after 1, 3, and 12 months and after 2 and 3 years.
2. PATIENTS AND METHODS
In a retrospective, quality study at Nykoebing Falster Hospital, Region Zealand, Denmark, we identified all major lower amputations, excluding those because of tumours and traumas, from January 1, 2010 to November 23, 2017. Before June 1, 2014, a classical gauze dressing, stump bandage, was used for patients undergoing these operations (control group), and from June 1, 2014, a new treatment with 5 days of postoperative CINPWT with “Prevena” was implemented to reduce complication rates (intervention group).12 During the period with CINPWT, 13 patients were considered unsuitable for this treatment; they received classical stump bandage, and in the analysis, they were included in the control group. The surgeon's argument in the medical records was lack of perfusion to the soft tissue, need for early wound check, necrotising fasciitis, or missing equipment.
The patients were treated by the same orthopaedic team at the Department of Orthopedics in Nykoebing Falster Hospital. Surgical treatment did not change in the period. The hospital is located in a region of Denmark with a population of low social class with high comorbidity. The hospital admission area was enlarged by the April 1, 2016, but this did not affect the composition of the patient population.
3. STATISTICS
3.1. Population data
Patients admitted twice or more during the study period because of multiple amputations were included on their first amputation. Revision surgery was defined as a surgical procedure in the operating room (OR) within 2 months from primary amputation. Length of stay was defined as postoperative length of stay in the hospital.
Differences between dichotomous variables in the two groups were pairwise analysed using a χ 2 test. A Shapiro–Wilk test of normality was used on continuous variables, that is, age, length of surgery, and length of stay. When the normality hypothesis could not be rejected (age), Welch's t‐test was used to compare the variables in the two groups; otherwise, a Mann‐Whitney U test was performed (length of surgery and stay).
3.2. Survival analysis
We constructed crude and age‐ and gender‐adjusted survival curves for a time period of 1300 days from first amputation, this being the time window where data from both groups were present. Survival curves and corresponding 95% confidence intervals were calculated using the Fleming–Harrington estimator. First, survival probabilities adjusted for gender and age at first operation were obtained by reweighting, using the overall age/gender distribution of the sample as a reference distribution.
P‐values were obtained by a log‐rank test (corrected for weighting in the adjusted case). A significance level of 0.05 was used. Second, survival probabilities were calculated for patients undergoing one amputation only. The reason for that is that the survival probability of a patient changes; it most likely increases when he/she undergoes a new surgery. Moreover, 12 patients had the first operation with the old treatment and the second operation with the new one. Data management, statistical analyses, and plots were carried out in R ver. 3.5.0, using the packages dplyr, survival, survey, and ggplot2.
3.3. Ethics
Implementation of the CINPWT was part of the quality improvement at Nykoebing Falster Orthopedic Department, and therefore, the study did not require approval from the Ethical Committee of the Zealand Region of Denmark. The local leader allowed the retrieval of data from medical records according to the Danish Health Care Act, chapter 9, § 42d, stk 2, nr. 2.
4. RESULTS
We included 309 patients undergoing 403 amputations. One patient was excluded because of a tumour, one because of a traumatic amputation, and two patients had errors in the registration.
A total of 139 patients received CINPWT, and 170 patients received stump bandage postoperatively (Table 1). A total of 192 patients were men, and 117 were women. Age at amputation was 73.3 years in the CINPWT group and 71.8 years in the bandage group (P = .23). 97 patients receiving CINPWT and 103 patients treated with stump bandage were amputated above knee (P = .12).
Table 1.
Patient undergoing major lower amputation in Nykoebing Falster Hospital, Denmark, January 1, 2010 to November 23, 2017, including patients with multiple amputations
| Gauze bandage (control) | CINPWT (intervention) | P‐value | Missing data (n) (control/intervention) | |
|---|---|---|---|---|
| N | 170 | 139 | ||
| Age (mean/median) | 71.8/71 | 73.3/74 | .23 | |
| Gender (M/F) | 105/65 | 87/52 | .98 | |
| Femoral amputation (KNFQ19) | 103 | 97 | .12 | |
| Crus amputation (KNGQ19) | 67 | 42 | .12 | |
| Indications: More than one possible | ||||
| Diabetes | 72 (42.4%) | 59 (42.4%) | >.99 | |
| Wound | 137 (80.6%) | 119 (85.6%) | .31 | |
| Atherosclerosis | 43 (25.2%) | 89 (64.0%) | <.0001 | |
| Pain | 71 (41.8%) | 87 (62.6%) | .0004 | |
| Dialysis | 16 (9.4%) | 15 (10.8%) | .83 | |
| Antithrombotic treatment | 72 | 104 | <.0001 | |
| Smokers/former smoker | 101 | 103 | .01 | 7/4 |
| Spinal anaesthesia | 9 | 19 | .02 | 4/4 |
| Surgical duration (min; mean/median) | 66.4/60 | 66.6/60 | .43 | 20/4 |
| Revision surgery | 27 | 17 | .45 | |
| Subsequent amputation | 38 | 28 | .74 | |
| Length of stay (d; mean/median) | 9.1/7 | 8.5/7 | .48 | |
CINPWT, closed incision negative pressure wound therapy.
Many patients had multiple indications for amputation. In both groups, 42.4% (P > .99) had diabetes, and 85.6% and 80.6%, respectively, had wounds (P = .31). More patients in the CINPWT groups than in the stump bandage group had atherosclerosis(P < .0001) or pain (P = .0004) as an indication. Antithrombotic treatment was more common in the CINPWT group than in the stump bandage group (P < .0001). Nineteen patients in the CINPWT group and nine patients in the stump bandage group had spinal anaesthesia (P = .02). Duration of surgery was equal, with a median duration of 60 minutes (P = .43). Smoking was frequent in both groups but more so in the CINPWT group, 75%, than in the stump bandage group, 59%, (P = .01).
Seventeen patients in the CINPWT group versus 27 in the stump bandage group needed revision surgery (P = .45); 28 in the CINPWT group and 38 in the control group had a subsequent amputation (P = .74). Length of stay for patients in the CINPWT group was 8.5 days and in the stump bandage group was 9.1 days (P = .48).
Fifteen patients in the CINPWT group and 14 patients in the stump bandage group died in hospital (P = .57). After 30 days, 22 patients in the CINPWT group and 30 patients in the stump bandage group had died (P = .79). After 90 days, 38 patients in the CINPWT group and 47 patients in the stump bandage group had died (P > .99). One year after amputation, 55 patients in the CINPWT group and 66 patients in the stump bandage group had died (P = .82). The survival probabilities adjusted for age and gender 2 years after amputation were .52 (.43‐.61) and .49 (.42‐.58), respectively, and 3 years after amputation were .36 (.25‐.50) and .39 (.32‐.47), respectively (Table 2).
Table 2.
Survival data, including all patients
| Gauze bandage (control) | CINPWT (intervention) | P‐value | Missing data (n) (control/intervention) | ||
|---|---|---|---|---|---|
| N | 170 | 139 | |||
| Mortality | In‐hospital | 14 (8.2%) | 15 (10.8%) | .57 | |
| 30‐day | 30 (17.6%) | 22 (15.8%) | .79 | ||
| 90‐day | 47 (27.6%) | 38 (27.3%) | >.99 | ||
| 1 year | 66 (38,8%) | 55 (39.6%) | .82 | 1/5 |
| Survival estimate (adjusted) | Gauze bandage (control) | CINPWT (intervention) | ||
|---|---|---|---|---|
| 1 year after amputation | 0.60 (0.53‐0.69) | 0.63 (0.55‐0.72) | Log‐rank test corrected for weighting—P‐value = .97 | |
| 2 years after amputation | 0.49 (0.42‐0.58) | 0.52 (0.43‐0.61) | ||
| 3 years after amputation | 0.39 (0.32‐0.47) | 0.36 (0.25‐0.50) | ||
| Survival estimate (unadjusted) | Gauze bandage (control) | CINPWT (intervention) | ||
|---|---|---|---|---|
| 1 year after amputation | 0.61 (0.54‐0.69) | 0.61 (0.53‐0.69) | Log‐rank test—P‐value = .58 | |
| 2 years after amputation | 0.50 (0.44‐0.59) | 0.49 (0.41‐0.58) | ||
| 3 years after amputation | 0.40 (0.33‐0.48) | 0.35 (0.26‐0.49) | ||
CINPWT, closed incision negative pressure wound therapy.
A total of 111 patients from the CINPWT group and 132 patients from the stump bandage group had one amputation only (Tables 3 and 4). The results for these subgroups of patients were similar to those for all patients.
Table 3.
Patient undergoing major lower amputation in Nykoebing Falster Hospital, Denmark, January 1, 2010 to November 23, 2017, excluding patients with multiple amputations
| Gauze bandage (control) | CINPWT (intervention) | P‐value | Missing data (n) (control/intervention) | |
|---|---|---|---|---|
| N | 132 | 111 | ||
| Age (mean/median) | 72.4/72 | 74.1/75 | .25 | |
| Gender (M/F) | 84/38 | 72/39 | .95 | |
| Femoral amputation (KNFQ19) | 88 | 83 | .22 | |
| Indications | ||||
| Diabetes | 59 (44.7%) | 46 (41.4%) | .7 | |
| Wound | 106 (80.3%) | 93 (83.8%) | .59 | |
| Atherosclerosis | 27 (20.4%) | 69 (62.1%) | <.0001 | |
| Pain | 53 (40.1%) | 69 (62.1%) | .001 | |
| Dialysis | 10 (7.6%) | 10 (9.0%) | .86 | |
| Antithrombotic treatment | 54 | 83 | <.0001 | |
| Smokers/former smokers | 73 | 78 | .03 | 7/4 |
| Spinal anaesthesia | 7 | 13 | .11 | 3/4 |
| Surgical duration (min) (mean/median) | 67.7/60 | 67.5/60 | .44 | 17/3 |
| Revision surgery | 10 | 10 | .86 | |
| Length of stay (days) (mean/median) | 8.4/7 | 8.0/7 | .75 | |
CINPWT, closed incision negative pressure wound therapy.
Table 4.
Survival data, excluding patients with multiple amputations
| Gauze bandage (control) | CINPWT (intervention) | P‐value | Missing data (n) (control/intervention) | ||
|---|---|---|---|---|---|
| N | 132 | 111 | |||
| Mortality | In‐hospital | 13 (9.8%) | 14 (12.6%) | .63 | |
| 30‐day | 28 (21.1%) | 21 (18.9%) | .78 | ||
| 90‐day | 40 (30.3%) | 34 (30.6%) | >.99 | ||
| 1 year | 54 (40.9%) | 48 (43.2%) | .62 | 1/5 |
| Survival estimate (adjusted) | Gauze bandage (control) | CINPWT (intervention) | ||
|---|---|---|---|---|
| 1 year after amputation | 0.58 (0.50‐0.68) | 0.60 (0.52‐0.70) | ||
| 2 years after amputation | 0.49 (0.41‐0.59) | 0.50 (0.41‐0.61) | Log‐rank test corrected for weighting—P‐value = .92 | |
| 3 years after amputation | 0.38 (0.30‐0.47) | 0.36 (0.25‐0.51) | ||
| Survival estimate (unadjusted) | Gauze bandage (control) | CINPWT (intervention) | ||
|---|---|---|---|---|
| 1 year after amputation | 0.59 (0.52‐0.68) | 0.57 (0.48‐0.67) | Log‐rank test—P‐value = .54 | |
| 2 years after amputation | 0.50 (0.42‐0.59) | 0.47 (0.39‐0.58) | ||
| 3 years after amputation | 0.39 (0.32‐0.48) | 0.36 (0.26‐0.49) | ||
CINPWT, closed incision negative pressure wound therapy.
A long‐rank test corrected for weighting showed no significant difference between the survival curves, both for the analysis including all patients (P = .97) (Figure 1) and for the analysis including the subgroup of patients with one amputation only (P = .92) (Figure 2).
Figure 1.
Survival curves; all patients included
Figure 2.
Survival curves; patients with multiple amputations excluded
5. DISCUSSION
In this comparison of patients undergoing major lower amputations from a hospital serving a well‐defined geographical area, we found no significant difference in revision surgery, in subsequent amputations, or in survival probabilities up till 3 years after amputation between patients for whom CINPWT and those for whom gauze bandage were used for wound dressing. The change from the old use of gauze bandage to the new use of CINPWT was not associated with a significant reduction in length of hospital stay, reflecting that accelerated discharge has been used widely in Denmark during the entire study period.
Since 2013, there has been a national guideline in Denmark for treatment of diabetic foot ulcers to prevent amputations,13, 14 but in our population, this has not changed the fraction of diabetic patients. We found significantly more patients with atherosclerosis and pain as indications for surgery in the CINPWT group than in the stump bandage group; we consider this information bias because of more detailed medical records and no true clinical phenomenon.
As a well‐known trend in the Danish population, the treatment with antithrombotics has increased by 9% during the last decade.15 The increase seen over the last 4 years is mainly for the consumption of novel oral anticoagulants. This is consistent with our data showing significantly more patients in the antithrombotic treatment group than in the CINPWT group. Another time trend reflected in our data is the development in the last decade of a significant rise in patients amputated using spinal anaesthesia to reduce postoperative complications and the need for opioid medication.16
A strength of our study was the inclusion of all consecutive patients over an 8‐year period. Another strength was the long follow‐up time, allowing for calculation of both 2‐year and 3‐year survival rates. The entire study population had a 2‐year mortality rate of around 50% and a 3‐year mortality rate of 60% to 65%. Our mortality rates after major non‐traumatic lower‐extremity amputation are comparable with the rest of Denmark as we present mortality rates of the better end of earlier published data.9, 10, 11
Patients with repeated revisions at the OR undergoing general anaesthesia are known to be at a significant risk of developing postoperative pulmonary complications and cardiac arrhythmias.17 Because of study design and missing details in medical journals, it was not possible in the present study to take minor revisions in the orthopaedic ward into account.
We had planned to use The American Society of Anesthesiologists Physical Status Scores (ASA‐Score) as a valuable indicator of comorbidity, but we found that these data were available only in the paper‐based part of the medical records. However, all included patients would have been classified as ASA scores 3 to 5, and the lack of ASA‐score is therefore unlikely to have affected our results.18 We searched for data of skin perfusion, distal blood pressure measurement, postoperative infections, and use of antibiotics, but these data were inconsistently recorded in the medical records.
Our study did not include the registration of costs. It is nevertheless possible to compare the estimated costs of the two wound dressing methods. The stump bandage price for one set of remedies runs up to 100 DKK/dressing (15 USD). In our study, gauze bandages were changed at day 2 and again at day 5. In reality, stump bandages are difficult to apply, and they often slip off the patient, and bandage changing is sometimes painful for the patient. Every change of dressing requires two nurses’ time and resources, which are also a considerable cost.
The CINPWT is more expensive in itself, costing 1000 DKK/treatment (150 USD). For most patients, the CINPWT can be left untouched the first 5 days. Figure 3 shows CINPWT applied postoperatively to a femoral stump, published with the patient's consent. It is the clinical experience at our department that the amputation stumps treated with CINPWT were less swollen and that further treatment with liner could be started earlier than for the patients with stump bandage.
Figure 3.
Closed incision negative pressure wound therapy (CINPWT) applied at a femoral stump
6. CONCLUSION
Neither the need for revision surgery and subsequent amputation nor the length of hospital stay and survival of patients undergoing non‐traumatic major lower amputation depended on bandage type. The use of CINPWT was a non‐inferior treatment compared with stump bandage. Major amputated patients have higher mortality rates at all points of time postoperatively than, for instance, hip fracture patients.19 There is an urgent need for randomised controlled trials of treatment options and disease management in this frail group of patients.
Stenqvist CP, Nielsen CT, Napolitano GM, et al. Does closed incision negative wound pressure therapy in non‐traumatic major lower‐extremity amputations improve survival rates? Int Wound J. 2019;16:1171–1177. 10.1111/iwj.13176
REFERENCES
- 1. Nam D, Sershon RA, Levine BR, Della Valle CJ. The use of closed incision negative‐pressure wound therapy in orthopaedic surgery. J Am Acad Orthop Surg. 2018;26(9):295‐302. [DOI] [PubMed] [Google Scholar]
- 2. Pachowsky M, Gusinde J, Klein A, et al. Negative pressure wound therapy to prevent seromas and treat surgical incisions after total hip arthroplasty. Int Orthop. 2012;36(4):719‐722. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Stannard JP, Volgas DA, McGwin G 3rd, et al. Incisional negative pressure wound therapy after high‐risk lower extremity fractures. J Orthop Trauma. 2012;26(1):37‐42. [DOI] [PubMed] [Google Scholar]
- 4. Willy C, Agarwal A, Andersen CA, et al. Closed incision negative pressure therapy: international multidisciplinary consensus recommendations. Int Wound J. 2017;14(2):385‐398. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Hyldig N, Birke‐Sorensen H, Kruse M, et al. Meta‐analysis of negative‐pressure wound therapy for closed surgical incisions. Br J Surg. 2016;103(5):477‐486. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Kotha V, Walter E, Stimac G, Kim P. Incisional application of negative pressure for nontraumatic lower extremity amputations: a review. Surg Technol Int. 2019;34:49‐55. [PubMed] [Google Scholar]
- 7. Gurney JK, Stanley J, York S, Rosenbaum D, Sarfati D. Risk of lower limb amputation in a national prevalent cohort of patients with diabetes. Diabetologia. 2018;61(3):626‐635. [DOI] [PubMed] [Google Scholar]
- 8. Jensen PS, Petersen J, Kirketerp‐Møller K, Poulsen I, Andersen O. Progression of disease preceding lower extremity amputation in Denmark: a longitudinal registry study of diagnoses, use of medication and healthcare services 14 years prior to amputation. BMJ Open. 2017;7(11):e016030. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Kristensen MT, Holm G, Kirketerp‐Møller K, Krasheninnikoff M, Gebuhr P. Very low survival rates after non‐traumatic lower limb amputation in a consecutive series: what to do? Interact Cardiovasc Thorac Surg. 2012;14(5):543‐547. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Kristensen MT, Holm G, Krasheninnikoff M, Jensen PS, Gebuhr P. An enhanced treatment program with markedly reduced mortality after a transtibial or higher non‐traumatic lower extremity amputation. Acta Orthop. 2016;87(3):306‐311. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Wied C, Kristensen MT, Tengberg PT, Holm G, Krasheninnikoff M, Troelsen A. Patients in need for major lower extremity amputations are a challenge. Ugeskr Laeger. 2015;177: V10140540. [PubMed] [Google Scholar]
- 12. Kristensen MT, Holm G, Gebuhr P. Difficult to predict early failure after major lower‐extremity amputations. Dan Med J. 2015;62(12):A5172. [PubMed] [Google Scholar]
- 13. Jørgensen ME, Almdal TP, Faerch K. Reduced incidence of lower‐extremity amputations in a Danish diabetes population from 2000 to 2011. Diabet Med. 2014;31(4):443‐447. [DOI] [PubMed] [Google Scholar]
- 14.National klinisk retningslinie for udredning og behandling af diabetiske fodsår. https://www.sst.dk/da/udgivelser/2013/~/media/54AAA6F78D8B4BF89929A4AFB8B68B0F.ashx. Accessed January 28, 2019.
- 15.Prab rapporten. http://www.dasaim.dk/wp-content/uploads/2017/01/PRAB-2016.pdf. Accessed January 28, 2019.
- 16. Niskakangas M, Dahlbacka S, Liisanantti J, Vakkala M, Kaakinen T. Spinal or general anaesthesia for lower‐limb amputation in peripheral artery disease ‐ a retrospective cohort study. Acta Anaesthesiol Scand. 2018;62(2):226‐233. [DOI] [PubMed] [Google Scholar]
- 17. Chery J, Semaan E, Darji S, Briggs WT, Yarmush J, D'Ayala M. Impact of regional versus general anesthesia on the clinical outcomes of patients undergoing major lower extremity amputation. Ann Vasc Surg. 2014;28(5):1149‐1156. [DOI] [PubMed] [Google Scholar]
- 18. Schmiegelow MT, Sode N, Riis T, Lauritzen JB, Duus BR, Lindberg‐Larsen M. Re‐amputations and mortality after below‐knee, through‐knee and above‐knee amputations. Dan Med J. 2018;65(12): A5520. [PubMed] [Google Scholar]
- 19. Stenqvist C, Madsen CM, Riis T, et al. Orthogeriatric service reduces mortality in patients with hip fracture. Geriatr Orthop Surg Rehabil. 2016;7(2):67‐73. [DOI] [PMC free article] [PubMed] [Google Scholar]